The present invention relates to an inductor having a high Q value for use in high frequency in a semiconductor integrated circuit (IC).
A conventional inductor will be described with reference to Figure 9. Referring to Figure 9, the reference numeral 1 denotes an inductor section, 2 denotes a drawing interconnect formed in the first layer, 3 denotes a drawing interconnect formed in the second layer, 5 denotes a connection between the first and second layers, 7 denotes an interlayer film, and 8 denotes a smoothing film.
That is, in the conventional inductor, the inductor section is constructed of a single layer and the second layer is used for the drawing interconnect for connection with other components.
As one of characteristics of an inductor, it is generally known that in order to obtain a large inductance value, the line length of the inductor must be increased.
With the above conventional construction, however, when the line length is increased in order to obtain a large inductance value, the serial resistance component increases due to the resistance of a wiring material constituting the inductor, resulting in lowering the Q value of the inductor.
Further, the increased line length of the inductor tends to increase the size of the entire inductor. EP-A-0 484 558 describes a high frequency inductor device comprising strip-like coil conductors formed on an insulating substrate. The inductor device includes a plurality of said strip-like coil conductors being arranged in plural layers and connected in parallel. The strip-like coil conductors are arranged such that an electric currents flowing through the conductors of the different layers has the same direction in corresponding conductor portions.
US-A-4 641 118 describes an electromagnet utilizing a coil consisting of a plurality of insulating substrates. On each substrate, a spiral conductor is formed. The spiral direction of conductors is reversed on each of successive substrates. The spiral centers and outer ends of successive conductors are connected in an alternating fashion, such that a serial connection of the conductors is obtained.
It is an object of the present invention to provide an improved high frequency inductor device.
This object is achieved by the features of claim 1, claim 3 and claim 4.
Further embodiments are subject-matter of dependent claims.
Embodiments of the present invention will be described with reference to the relevant drawings.
Figure 1 shows the first embodiment of the high-Q inductor for high frequency useful for the understanding of the present invention. Referring to Figure 1, the reference numeral 11 denotes a meander-type first-layer inductor section (the "inductor section" as used herein corresponds to an "inductor element" to be recited in the claims), 12 and 13 denote first-layer drawing interconnects, 14 denotes a second-layer inductor section, 15 and 16 denote connections between the first and second layers, 17 denotes an interlayer film, and 18 denotes a smoothing film.
Each of the connections 15 and 16 is composed of nine contact portions each having a size of about 1 µm square, for example.
In this embodiment, therefore, the inductor section, which is conventionally constructed using only one layer, is of a two-layer structure where two inductor sections are formed in the first and second layers and connected in parallel with each other.
The above construction makes it possible to obtain a high Q-value inductor for high frequency which overcomes the conventional problem of having a large serial resistance component in low frequency and high frequency and thus a lowered Q value, by increasing the cross section and suppressing lowering of the Q value which otherwise occurs due to a skin effect in high frequency.
It should be noted that the first and second layers may be connected in parallel over the entire inductor sections. This construction is also included in the present invention.
Figure 2 shows the second embodiment of the high-Q inductor for high frequency useful for the understanding of the present invention. Referring to Figure 2, the reference numeral 21 denotes a spiral-shaped first-layer inductor section, 22 denotes a first-layer drawing interconnect, 23 denotes a spiral-shaped second-layer inductor section, 24 denotes a drawing interconnect from the second-layer inductor section 23 formed in the third layer, 25 and 26 denote connections between the first and second layers, 27 and 28 denote interlayer films, 29 denotes a smoothing film, and 210 denotes a connection between the second and third layers. The first-layer inductor section 22 and the second-layer inductor section 23 are spiraled in the same direction.
In this embodiment, therefore, the inductor section, which is conventionally constructed using only one layer, is of a two-layer structure where the inductor sections22 and 23 are respectively formed in the first and second layers and connected in parallel with each other. This construction makes it possible to obtain a high Q-value inductor for high frequency which overcomes the conventional problem of having a large serial resistance component in low frequency and high frequency and thus a lowered Q value, by increasing the cross section and suppressing lowering of the Q value which otherwise occurs due to a skin effect in high frequency.
It should be noted that the first and second layers may be connected in parallel over the entire inductor sections. This construction is also included in the present invention.
In this embodiment, the three-layer inductor was exemplified. It is also possible to construct a similar structure composed of four or more layers with a drawing interconnect being formed in the bottom layer.
Figure 3 shows the third embodiment of the high-Q inductor for high frequency according to the present invention. Referring to Figure 3, the reference numeral 31 denotes a spiral-shaped first-layer inductor section, 32 denotes a first-layer drawing interconnect, 33 denotes a spiral-shaped second-layer inductor section, 34 denotes a second-layer drawing interconnect, 35 denotes connections between the first and second layers, 37 denotes an interlayer film, and 38 denotes a smoothing film.
The first and second inductor sections 31 and 33 are connected in parallel with each other.
Embodiment 3 is characterized in that the second-layer drawing interconnect 34 is formed using the layer in which the second-layer inductor section 33 is formed. In order to prevent the second-layer inductor section 33 from being in contact with the drawing interconnect 34 in the same layer, the second-layer inductor section 33 is cut off at the positions where the drawing interconnect 34 crosses. The cut-off ends of the inductor section 33 are connected with the first-layer inductor section 31 via the connections 35. By this construction, the second-layer inductor section 33 can serve as one substantially spiral-shaped inductor section.
In this embodiment, therefore, the inductor section, which is conventionally constructed using only one layer, is of a two-layer structure where inductor sections are formed in the first and second layers and connected in parallel with each other. Furthermore, the inductor sections are formed in the layers in which the drawing interconnects are formed. As a result, it is possible, even in a process where a smaller number of wiring layers are used, to obtain a high Q-value inductor for high frequency which overcomes the conventional problem of having a large serial resistance component in low frequency and high frequency and thus a lowered Q value, by increasing the cross section and suppressing lowering of the Q value which otherwise occurs due to a skin effect in high frequency.
Thus, Embodiment 3 is characterized in that one of the drawing interconnects is formed using the wiring layer for the inductor section, which is different from Embodiment 2 where the layer for forming the drawing interconnect is separately provided.
It should be noted that the first and second layers may be connected in parallel over the entire inductor sections. This construction is also included in the present invention.
In this embodiment, the two-layer inductor was exemplified. It is also possible to construct a similar structure composed of three or more layers with a drawing interconnect being formed in any of the layers. In this case, portions of an inductor section at which the drawing interconnect crosses can be connected with an adjacent upper or lower inductor section.
Figures 7 and 8 are graphs showing comparison of performances of the two-layer inductor according to the present invention and a conventional one-layer inductor.
Figure 7 is a graph obtained by plotting a variation of the resistance (R) with respect to the length (L). It is observed from this figure that R is smaller in the two-layer inductor according to the present invention.
Figure 8 is a graph obtained by plotting a variation of the Q value (Q) with respect to the length (L). It is observed from this figure that Q is greater in the two-layer inductor according to the present invention.
Figure 4 shows the fourth embodiment of the high-Q inductor for high frequency according to the present invention. Referring to Figure 4, the reference numeral 41 denotes a spiral-shaped first-layer inductor section, 42 denotes a first-layer drawing interconnect, 43 denotes a connection between the first and second layers, 44 denotes a spiral-shaped second-layer inductor section, 45 denotes a connection between the second and third layers, 46 denotes a spiral-shaped third-layer inductor section, 47 denotes a connection between the third and fourth layers, 48 denotes a spiral-shaped fourth-layer inductor section, 49 denotes a fourth-layer drawing interconnect, 410, 411, and 412 denote interlayer films, and 413 denotes a smoothing film.
In this embodiment, the adjacent inductor sections are connected with each other. Specifically, the centers or the outer ends of the adjacent inductor sections are connected with each other. These inductor sections are therefore connected in series with each other.
In this embodiment, the second-layer and fourth-layer inductor sections have a shape inverted upside down from that of the first-layer and third-layer inductor sections. By this arrangement, the directions of the magnetic fields generated by the respective inductor sections are the same, resulting in effective coupling.
In the conventional structure where the inductor section is constructed using only a single layer, when the entire length of the inductor section is increased to obtain a high Q value, the size of the inductor section also increases. On the contrary, in Embodiment 4, since the length of the inductor sections is increased stereoscopically as a whole, the resultant size is compact.
The four-layer structure was described in this embodiment. However, as shown in Figure 5, the number of layers may be increased to five or six, for example, in a similar structure. The structure is simpler when the number of layers is even, because the drawing interconnect can be formed to be connected with the outer end of the bottom inductor section.
When the number of layers is odd, the drawing interconnect can be arranged in a manner described in Figure 2 or 3.
Alternatively, as shown in Figure 6, a pair of adjacent inductor sectors may have the same spiral direction, and adjacent pairs of adjacent inductor sectors may have different spiral directions. In this case, one inductor sector of one pair is connected with one of another pair as shown in Figure 6 so that all the inductor sectors are serially connected.
In the above case, also, the directions of the magnetic fields generated by the respective inductor sectors are the same, resulting in effective coupling.
Thus, according to the present invention, the inductor section, which is conventionally constructed of a single wiring layer, is of a multi-layer structure. As a result, a high Q-value inductor which has a reduced serial resistance component and is free from an influence of a skin effect can be fabricated in an IC.
Many modifications and variations of the present invention will be apparent to those skilled in the art without departing from the scope of the invention, as defined by the appended claims.